Phosphorylation of proteins is a fundamental mechanism for regulation of many cellular processes. Although protein phosphorylation at serine and threonine residues is quantitatively dominating in eukaryotic cells, reversible tyrosine phosphorylation seems to play a pivotal role in regulation of cell growth and differentiation as well as in neoplastic transformation (Hunter, Cell 80: 225-236 (1995); Schlessinger and Ullrich, Neuron 9: 383-391 (1992); Cantley et al., Cell 64: 281-302 (1991); Ullrich and Schlessinger, Cell 61:203-212(1990); Hunter, Curr. Opin. Cell. Biol. 1:1168-1181 (1989)); Hunter and Cooper, Annu. Rev. Biochem. 54: 897-930 (1985)).
The regulation of protein tyrosine phosphorylation in vivo is mediated by the opposing actions of protein tyrosine kinases (PTKs) and protein tyrosine phosphatases (PTPases). The level of protein tyrosine phosphorylation of cellular proteins is determined by the balanced activities of PTKs and PTPase (Hunter, 1995, supra).
PTPases--an overview
The protein phosphatases are composed of at least two separate and distinct families (Hunter, T., Cell 58: 1013-1016 (1989)) the protein serine/threonine phosphatases and PTPases.
The PTPases are a family of enzymes that can be classified into two groups: a) intracellular or nontransmembrane PTPases and b) receptor-type or transmembrane PTPases.
Intracellular PTPases: All known intracellular type PTPases contain a single conserved catalytic phosphatase domain consisting of 220-240 amino acid residues. The regions outside the PTPase domains are believed to play important roles in localizing the intracellular PTPases subcellularly (Mauro, L. J. and Dixon, J. E. TIBS 19: 151-155 (1994)). The first intracellular PTPase to be purified and characterized was PTP1B which was isolated from human placenta (Tonks et al., J. Biol. Chem. 263: 6722-6730 (1988)). Shortly after, PTP1B was cloned (Charbonneau et al., Proc. Natl. Acad. Sci. U.S.A. 86: 5252-5256 (1989); Chernoff et al., Proc. Natl. Acad. Sci. U.S.A. 87: 2735-2789 (1989)). Other examples of intracellular PTPases include (1) T-cell PTPase (Cool et al. Proc. Natl. Acad. Sci. U.S.A. 86: 5257-5261 (1989)), (2) rat brain PTPase (Guan et al., Proc. Natl. Acad. Sci. U.S.A. 87:1501-1502 (1990)), (3) neuronal phosphatase STEP (Lombroso et al., Proc. Natl. Acad. Sci. U.S.A. 88: 7242-7246 (1991)), (4) ezrin-domain containing PTPases: PTPMEG1 (Guet al., Proc. Natl. Acad. Sci. U.S.A. 88: 5867-57871 (1991)), PTPH1 Yang and Tonks, Proc. Natl. Acad. Sci. U.S.A. 88: 5949-5953 (1991), PTPD1 and PTPD2 (M.o slashed.ller et al., Proc. Natl. Acad. Sci. U.S.A. 91: 7477-7481 (1994)), FAP-1/BAS (Sato et al., Science 268: 411-415 (1995); Banville et al., J. Biol. Chem. 269: 22320-22327 (1994); Maekawa et al., FEBS Letter 337: 200-206 (1994)), and SH2 domain containing PTPases: PTP1C/SH-PTP1 (Plutzky et al., Proc. Natl. Acad. Sci. U.S.A. 89:1123-1127 (1992); Shen et al., Nature Lond. 352:736-739 (1991)) and PTP1D/Syp/SH-PTP2 (Vogel et al., Science 259: 1611-1614 (1993); Feng et al., Science 259:1607-1611(1993); Bastein et al., Biochem. Biophys. Res. Comm. 196: 124-133 (1993)).
Low molecular weight phosphotyrosine-protein phosphatase (LMW-PTPase) shows very little sequence identity to the intracellular PTPases described above. However, this enzyme belongs to the PTPase family due to the following characteristics: (i) it possesses the PTPase active site motif: Cys-Xxx-Xxx-Xxx-Xxx-Xxx-Arg (SEQ ID NO:1) (Cirri et al., Eur. J. Biochem. 214: 647-657 (1993)); (ii) this Cys residue forms a phospho-intermediate during the catalytic reaction similar to the situation with `classical` PTPases (Cirri et al, supra; Chiarugi et al., FEBS Lett. 310: 9-12 (1992)); (iii) the overall folding of the molecule shows a surprising degree of similarity to that of PTP1B and Yersinia PTP (Su et al., Nature 370: 575-578 (1994)).
Receptor-type PTPases consist of a) a putative ligand-binding extracellular domain, b) a transmembrane segment, and c) an intracellular catalytic region. The structures and sizes of the putative ligand-binding extracellular domains of receptor-type PTPases are quite divergent. In contrast, the intracellular catalytic regions of receptor-type PTPases are very homologous to each other and to the intracellular PTPases. Most receptor-type PTPases have two tandemly duplicated catalytic PTPase domains.
The first receptor-type PTPases to be identified were (1) CD45ILCA (Ralph, S. J., EMBO J. 6:1251-1257 (1987)) and (2) LAR (Streuli et al., J. Exp. Med. 168: 1523-1530 (1988)) that were recognized to belong to this class of enzymes based on homology to PTP1B (Charbonneau et al., Proc. Natl. Acad. Sci. U.S.A. 86: 5252-5256 (1989)). CD45 is a family of high molecular weight glycoproteins and is one of the most abundant leukocyte cell surface glycoproteins and appears to be exclusively expressed upon cells of the hematopoietic system (Trowbridge and Thomas, Ann. Rev. Immunol. 12: 85-116 (1994)).
The identification of CD45 and LAR as members of the PTPase family was quickly followed by identification and cloning of several different members of the receptor-type PTPase group. Thus, 5 different PTPases, (3) PTP.alpha., (4) PTP.beta., (5) PTP.delta., (6) PTP.epsilon., and (7) PTP.zeta., were identified in one early study (Krueger et al., EMBO J. 9: 3241-3252 (1990)). Other examples of receptor-type PTPases include (8) PTP.gamma. (Barnea et al., Mol. Cell. Biol. 13:1497-1506 (1995)) which, like PTP.zeta. (Krueger and Saito, Proc. Natl. Acad. Sci. U.S.A. 89: 7417-7421 (1992)) contains a carbonic anhydrase-like domain in the extracellular region, (9) PTP.mu. (Gebbink et al., FEBS Letters 290: 123-130 (1991), (10) PTP.kappa. (Jiang et al., Mol. Cell. Biol. 13:2942-2951 (1993)). Based on structural differences the receptor-type PTPases may be classified into subtypes (Fischer et al., Science 253: 401-406 (1991)): (I) CD45; (II) LAR, PTP.delta., (11) PTP.sigma.; (III) PTP.beta., (12) SAP-1 (Matozaki et al., J. Biol. Chem. 269: 2075-2081 (1994)), (13) PTP-U2/GLEPP1 (Seimiya et al., Oncogene 10: 1731-1738 (1995); (Thomas et al., J. Biol. Chem. 269:19953-19962 (1994)), and (14) DEP-1; (IV) PTP.alpha.,.sub.-- PTP.epsilon.. All receptor-type PTPases except Type IV contain two PTPase domains. Novel PTPases are continously identified, and it is anticipated that more than 500 different species will be found in the human genome, i.e. close to the predicted size of the protein tyrosine kinase superfamily (Hanks and Hunter, FASEB J. 9: 576-596 (1995)).
PTPases are the biological counterparts to protein tyrosine kinases (PTKs). Therefore, one important function of PTPases is to control, down-regulate, the activity of PTKs. However, a more complex picture of the function of PTPases now emerges. Several studies have shown that some PTPases may actually act as positive mediators of cellular signaling. As an example, the SH2 domain-containing PTP1D seems to act as a positive mediator in insulin-stimulated Ras activation (Noguchi et al., Mol. Cell. Biol. 14: 6674-6682 (1994)) and of growth factor-induced mitogenic signal transduction (Xiao et al., J. Biol. Chem. 269: 21244-21248 (1994)), whereas the homologous PTP1 C seems to act as a negative regulator of growth factor-stimulated proliferation (Bignon and Siminovitch, Clin. Immunol. Immunopathol. 73:168-179 (1994)). Another example of PTPases as positive regulators has been provided by studies designed to define the activation of the Src-family of tyrosine kinases. In particular, several lines of evidence indicate that CD45 is positively regulating the activation of hematopoietic cells, possibly through dephosphorylation of the C-terminal tyrosine of Fyn and Lck (Chan et al., Annu. Rev. Immunol. 12: 555-592 (1994)).
Dual specificity protein tyrosine phosphatases (dsPTPases) define a subclass within the PTPases family that can hydrolyze phosphate from phosphortyrosine as well as from phosphor-serine/threonine. dsPTPases contain the signature sequence of PTPases: His-Cys-Xxx-Xxx-Gly-Xxx-Xxx-Arg (SEQ ID NO:2). At least three dsPTPases have been shown to dephosphorylate and inactivate extracellular signal-regulated kinase (ERKs)/mitogen-activated protein kinase (MAPK): MAPK phosphatase (CL100, 3CH134) (Charles et al., Proc. Natl. Acad. Sci. U.S.A. 90: 5292-5296 (1993)); PAC-1 (Ward et al., Nature 367: 651-654 (1994)); rVH6 (Mourey et al., J. Biol. Chem. 271: 3795-3802 (1996)). Transcription of dsPTPases are induced by different stimuli, e.g. oxidative stress or heat shock (Ishibashi et al., J. Biol. Chem. 269: 29897-29902 (1994); Keyse and Emslie, Nature 359: 644-647 (1992)). Further, they may be involved in regulation of the cell cycle: cdc25 (Millar and Russell, Cell 68: 407-410 (1992)); KAP (Hannon et al., Proc. Natl. Acad. Sci. U.S.A. 91: 1731-1735 (1994)). Interestingly, tyrosine dephosphorylation of cdc2 by a dual specific phosphatase, cdc25, is required for induction of mitosis in yeast (review by Walton and Dixon, Annu. Rev. Biochem. 62: 101-120 (1993)).
PTPase specificity
Several studies have addressed the question of PTPase specificity using synthetic peptides and provided important insight with respect to primary structural sequence requirements for substrate recognition (Ramachandran et al., Biochemistry 31: 4232-4238 (1992); Cho, H. et al., Biochemistry 31: 133-138 (1992); Zhang, Z-Y. et al., Proc. Natl. Acad. Sci. U.S.A. 90, 4446-4450 (1993); Zhang, Z.-Y. et al., Biochemistry 33: 2285-2290 (1994)). However, an obvious limitation of this approach is the lack of defined three-dimensional structure of the peptide analogs. Likewise, the PTPases utilized for these analyses are removed from their natural environment. Since at least part of the PTPase specificity seems to be conveyed by a defined subcellular localization (Mauro and Dixon, TIBS 19: 151-155 (1994)), it is essential that such studies are complemented with measurements of PTPase activity towards cellular substrates in intact cells.
Phosphotyrosine recognition in signal transduction
Hormones, growth factors, cytokines, antigens, extracellular matrix components as well as molecules positioned at the cell surface induce signal transduction by binding to specific cell surface structures or receptors on target cells (reviewed in Pawson, Nature 373: 573-580 (1995)). The resulting cellular signal is often mediated through a series of phosphorylation and dephosphorylation reactions on tyrosine residues of signaling molecules. To allow efficient and selective signaling, several recognition units for phosphotyrosine (pTyr) have developed during evolution: a) PTPases; b) Src-homology-2 (SH2) domains; c) pTyr-binding (PTB) domains. As described above, the recognition of pTyr by PTPases leads to dephosphorylation with concommitant dissociation from the molecular target. Dephosphorylation may either lead to upregulation or downregulation of the signal. In contrast, SH2 domains and PTB domains primarily act as docking molecules with little or no catalytic activity. In other words, tyrosine phosphorylated proteins have the capacity to bind other proteins containing SH2 domains or PTB domains thereby controlling the subcellular location of signaling molecules. There appears to be a significant degree of selectivity in SH2 domain recognition of pTyr and their surroundings. Thus, SH2 domains from the Src kinase family bind the peptide pTyr-Glu-Glu-Ile (SEQ ID NO:4) in a relatively selective manner, whereas the PTPD1 seems to recognize at least five, primarily hydrophobic residues C-terminal to the pTyr (Pawson, supra). Certain PTPase domains, in particular the C-terminal domain of some receptor-type PTPases, seem to have little or no catalytic activity. It may be hypothesized that these domains have a function as pTyr recognition units similar to SH2 domains and PTB domains. Inhibition of signal transduction processes could, in principle, be achieved by using non-hydrolyzable pTyr-containing peptides with preferential affinity for specific PTPases, SH2 domains or PTB domains. However, due to the lack of efficient bioavailability of peptides there is a need for development of either peptidomimetics or novel small molecules with preferential binding to pTyr recognition units of specific cellular targets. Such selective compounds can either initiate, increase or decrease defined signal transduction processes.
PTPases: Inhibitors
In an early study, vanadate was found to inhibit protein-tyrosine phosphatases in mammalian cells with a concomitant increase in the level of phosphotyrosine in cellular proteins leading to transformation (Klarlund, Cell 41: 707-717 (1985)). Vanadium-based phosphatase inhibitors are relatively unspecific. Therefore, to assess the importance of specific structures on PTPase activity more selective inhibitors are needed. One possibility for obtaining selective PTPase inhibitors would be through design of different ancillary ligands for peroxovanadium-based compounds (Posner et al., J. Biol. Chem. 269:4596-4604 (1994)). Another avenue taken by several investigators has been to incorporate nonhydrolyzable tyrosine phosphate analogs into specific peptide substrates: (1) phosphonomethyl phenylalanine (Zhang et al., Biochemistry 33: 2285-2290 (1994)); (2) difluorophosphono-methyl phenylalanine Burk et al., Synthesis 11: 1019-1020 (1991)); (3) L-O-malonyltyrosine (Kole et al., Biochem. Biophys. Res. Commun. 209: 817-822 (1995)); (4) cinnamic acid (Moran et al., J. Am. Chem. Soc. 117: 10787-10788 (1995); Cao et al., Bioorganic Med. Chem. Lett. 5: 2953-2958 (1995)); (5) sulfotyrosyl (Liotta et al., J. Biol. Chem. 269: 22996-23001 (1994)). A surprising degree of selectivity is observed with simple peptide analogs containing phosphonodifluoromethyl phenylalanine as a substitute for tyrosine (Chen et al., Biochem. Biophys. Res. Commun. 216: 976-984 (1995)). Important information has further been obtained with synthetic peptides containing sulfotyrosyl residues. A synthetic peptide corresponding to the amino acid sequence of a defined loop of the insulin receptor tyrosine kinase, Thr-Arg-Asp-Ile-Xxx-Glu-Thr-Asp-Xxx-Xxx-Arg-Lys (SEQ ID NO:3) (where Xxx denotes sulfotyrosyl), acts as a PTPase inhibitor (Liotta et al., 1994, supra). More importantly, this peptide, when tagged with stearic acid can penetrate cells, and stimulate the action of insulin (Liotta et al., 1994, supra).
PTPases: the insulin receptor signaling pathway/diabetes
Insulin is an important regulator of different metabolic processes and plays a key role in the control of blood glucose. Defects related to its synthesis or signaling lead to diabetes mellitus. Binding of insulin to its receptor causes rapid (auto)phosphorylation of several tyrosine residues in the intracellular part of the .beta.-subunit. Three closely positioned tyrosine residues (the tyrosine-1150 domain) must all be phosphorylated to obtain full activity of the insulin receptor tyrosine kinase (IRTK) which transmits the signal further downstream by tyrosine phosphorylation of other cellular substrates, including insulin receptor substrates (IRS-1) (Wilden et al., J. Biol. Chem. 267: 16660-16668 (1992); Myers and White, Diabetes 42: 643-650 (1993); Lee and Pilch, Am. J. Physiol. 266: C319-C334 (1994); White et al., J. Biol. Chem. 263: 2969-2980 (1988)). The structural basis for the function of the tyrosine-triplet has been provided by recent X-ray crystallographic studies of IRTK that showed tyrosine-1150 to be autoinhibitory in its unphosphorylated state (Hubbard et al., Nature 372: 746-754 (1994)).
Several studies clearly indicate that the activity of the auto-phosphorylated IRTK can be reversed by dephosphorylation in vitro (reviewed in Goldstein, Receptor 3: 1-15 (1993); Mooney and Anderson, J. Biol. Chem. 264: 6850-6857 (1989)), with the tri-phosphorylated tyrosine-1150 domain being the most sensitive target for protein-tyrosine phosphatases (PTPases) as compared to the di- and mono phosphorylated forms (King et al., Biochem. J. 275: 413-418 (1991)). It is, therefore, tempting to speculate that this tyrosine-triplet functions as a control switch of IRTK activity. Indeed, the IRTK appears to be tightly regulated by PTP-mediated dephosphorylation in vivo (Khan et al., J. Biol. Chem. 264: 12931-12940 (1989); Faure et al., J. Biol. Chem. 267: 11215-11221 (1992); Rothenberg et al., J. Biol. Chem. 266: 8302-8311 (1991)). The intimate coupling of PTPases to the insulin signaling pathway is further evidenced by the finding that insulin differentially regulates PTPase activity in rat hepatoma cells (Meyerovitch et al., Biochemistry 31: 10338-10344 (1992)) and in livers from alloxan diabetic rats (Boylan et al., J. Clin. Invest. 90: 174-179 (1992)).
Relatively little is known about the identity of the PTPases involved in IRTK regulation. However, the existence of PTPases with activity towards the insulin receptor can be demonstrated as indicated above. Further, when the strong PTPase-inhibitor pervanadate is added to whole cells an almost full insulin response can be obtained in adipocytes (Fantus et al., Biochemistry 28: 8864-8871 (1989); Eriksson et al., Diabetologia 39: 235-242 (1995)) and skeletal muscle (Leighton et al., Biochem. J. 276: 289-292 (1991)). In addition, recent studies show that a new class of peroxovanadium compounds act as potent hypoglycemic compounds in vivo (Posner et al.,supra). Two of these compounds were demonstrated to be more potent inhibitors of dephosphorylation of the insulin receptor than of the EGF-receptor.
It was recently found that the ubiquitously expressed SH2 domain containing PTPase, PTP1D (Vogel et al., 1993, supra), associates with and dephosphorylates IRS-1, but apparently not the IR itself (Kuhne et al., J. Biol. Chem. 268:11479-11481 (1993); (Kuhne et al., J. Biol. Chem. 269:15833-15837 (1994)).
Previous studies suggest that the PTPases responsible for IRTK regulation belong to the class of membrane-associated (Faure et al., J. Biol. Chem. 267: 11215-11221 (1992)) and glycosylated molecules (Haring et al., Biochemistry 23: 3298-3306 (1984); Sale, Adv. Prot. Phosphatases 6:159-186 (1991)). Hashimoto et al. have proposed that LAR might play a role in the physiological regulation of insulin receptors in intact cells (Hashimoto et al., J. Biol. Chem. 267: 13811-13814 (1992)). Their conclusion was reached by comparing the rate of dephosphorylation/inactivation of purified IR using recombinant PTP1B as well as the cytoplasmic domains of LAR and PTP.alpha.. Antisense inhibition was recently used to study the effect of LAR on insulin signaling in a rat hepatoma cell line (Kulas et al., J. Biol. Chem. 270: 2435-2438 (1995)). A suppression of LAR protein levels by about 60 percent was paralleled by an approximately 150 percent increase in insulin-induced auto-phosphorylation. However, only a modest 35 percent increase in IRTK activity was observed, whereas the insulin-dependent phosphatidylinositol 3-kinase (PI 3-kinase) activity was significantly increased by 350 percent. Reduced LAR levels did not alter the basal level of IRTK tyrosine phosphorylation or activity. The authors speculate that LAR could specifically dephosphoylate tyrosine residues that are critical for PI 3-kinase activation either on the insulin receptor itself or on a downstream substrate.
While previous reports indicate a role of PTPA in signal transduction through src activation (Zheng et al., Nature 359: 336-339 (1992); den Hertog et al., EMBO J. 12: 3789-3798 (1993)) and interaction with GRB-2 (den Hertog et al., EMBO J. 13: 3020-3032 (1994); Su et al., J. Biol. Chem. 269:18731-18734 (1994)), a recent study suggests a function for this phosphatase and its close relative PTP.epsilon. as negative regulators of the insulin receptor signal (M.o slashed.ller et al., 1995 supra). This study also indicates that receptor-like PTPases play a significant role in regulating the IRTK, whereas intracellular PTPases seem to have little, if any, activity towards the insulin receptor. While it appears that the target of the negative regulatory activity of PTPases .alpha. and .epsilon. is the receptor itself, the downmodulating effect of the intracellular TC-PTP seems to be due to a downstream function in the IR-activated signal. Although PTP1B and TC-PTP are closely related, PTP1B had only little influence on the phosphorylation pattern of insulin-treated cells. Both PTPases have distinct structural features that determine their subcellular localization and thereby their access to defined cellular substrates (Frangione et al., Cell 68: 545-560 (1992); Faure and Posner, Glia 9: 311-314 (1993)). Therefore, the lack of activity of PTP1B and TC-PTP towards the IRTK may, at least in part, be explained by the fact that they do not co-localize with the activated insulin receptor. In support of this view, PTP1B and TC-PTP have been excluded as candidates for the IR-associated PTPases in hepatocytes based on subcellular localization studies (Faure et al., J. Biol. Chem. 267: 11215-11221 (1992)).
The transmembrane PTPase CD45, which is believed to be hematopoietic cell-specific, was in a recent study found to negatively regulate the insulin receptor tyrosine kinase in the human multiple myeloma cell line U266 (Kulas et al., J. Biol. Chem. 271: 755-760 (1996)).
PTPases: somatostatin
Somatostatin inhibits several biological functions including cellular proliferation (Lamberts et al., Molec. Endocrinol. 8:1289-1297 (1994)). While part of the antiproliferative activities of somatostatin are secondary to its inhibition of hormone and growth factor secretion (e.g. growth hormone and epidermal growth factor), other antiproliferative effects of somatostatin are due to a direct effect on the target cells. As an example, somatostatin analogs inhibit the growth of pancreatic cancer presumably via stimulation of a single PTPase, or a subset of PTPases, rather than a general activation of PTPase levels in the cells (Liebow et al., Proc. Natl. Acad. Sci. U.S.A. 86: 2003-2007 (1989); Colas et al., Eur. J. Biochem. 207: 1017-1024 (1992)). In a recent study it was found that somatostatin stimulation of somatostatin receptors SSTR1, but not SSTR2, stably expressed in CHO-K1 cells can stimulate PTPase activity and that this stimulation is pertussis toxin-sensitive. Whether the inhibitory effect of somatostatin on hormone and growth factor secretion is caused by a similar stimulation of PTPase activity in hormone producing cells remains to be determined.
PTPases: the immune system/autoimmunity
Several studies suggest that the receptor-type PTPase CD45 plays a critical role not only for initiation of T cell activation, but also for maintaining the T cell receptor-mediated signaling cascade. These studies are reviewed in: Weiss A, Ann. Rev. Genet. 25: 487-510 (1991); Chan et al., Annu. Rev. Immunol. 12: 555-592 (1994); Trowbridge and Thomas, Annu. Rev. Immunol. 12: 85-116 (1994).
The exact function of CD45 in lymphocyte activation is currently under intense investigation in many laboratories. Several studies suggest that the PTPase activity of CD45 plays a role in the activation of Lck, a lymphocyte-specific member of the Src family protein-tyrosine kinase (Mustelin et al., Proc. Natl. Acad. Sci. U.S.A. 86: 6302-6306 (1989); Ostergaard et al., Proc. Natl. Acad. Sci. U.S.A. 86: 8959-8963 (1989)). These authors hypothesized that the phosphatase activity of CD45 activates Lck by dephosphorylation of a C-terminal tyrosine residue, which may, in turn, be related to T-cell activation. In a recent study it was found that recombinant p56.sup.lck specifically associates with recombinant CD45 cytoplasmic domain protein, but not to the cytoplasmic domain of the related PTP.alpha. (Ng et al., J. Biol. Chem. 271: 1295-1300 (1996)). The p56.sup.lck -CD45 interaction seems to be mediated via a nonconventional SH2 domain interaction not requiring phosphotyrosine. In immature B cells, another member of the Src family protein-tyrosine kinases, Fyn, seems to be a selective substrate for CD45 compared to Lck and Syk (Katagiri et al., J. Biol. Chem. 270: 27987-27990 (1995)).
HePTP, a hematopoietic cell specific PTPase, is induced after activation of resting T cells and may play a role in late T cell activation or as a negative regulator of T cell responses (Zanke et al, Eur. J. Immunol. 22: 235-239 (1992)). Likewise, the hematopoietic cell specific PTP1C seems to act as a negative regulator and play an essential role in immune cell development. In accordance with the above-mentioned important function of CD45, HePTP and PTP1C, selective PTPase inhibitors may be attractive drug candidates both as immunosuppressors and as immunostimulants. One recent study illustrates the potential of PTPase inhibitors as immunmodulators by demonstrating the capacity of the vanadium-based PTPase inhibitor, BMLOV, to induce apparent B cell selective apoptosis compared to T cells (Schieven et al., J. Biol. Chem. 270: 20824-20831 (1995)).
PTPases: cell-cell interactions/cancer
Focal adhesion plaques, an in vitro phenomenon in which specific contact points are formed when fibroblasts grow on appropriate substrates, seem to mimic, at least in part, cells and their natural surroundings. Several focal adhesion proteins are phosphorylated on tyrosine residues when fibroblasts adhere to and spread on extracellular matrix (Gumbiner, Neuron 11, 551-564 (1993)). However, aberrant tyrosine phosphorylation of these proteins can lead to cellular transformation. The intimate association between PTPases and focal adhesions is supported by the finding of several intracellular PTPases with ezrin-like N-terminal domains, e.g. PTPMEG1 (Gu et al., Proc. Natl. Acad. Sci. U.S.A. 88: 5867-5871 (1991)), PTPH1 (Yang and Tonks, Proc. Natl. Acad. Sci. U.S.A. 88: 5949-5953 (1991)) and PTPD1 (M.o slashed.ller et al., Proc. Natl. Acad. Sci. U.S.A. 91: 7477-7481(1994)). The ezrin-like domain show similarity to several proteins that are believed to act as links between the cell membrane and the cytoskeleton. PTPD1 was found to be phosphorylated by and associated with c-src in vitro and is hypothesized to be involved in the regulation of phosphorylation of focal adhesions (M.o slashed.ller et al., supra).
PTPases may oppose the action of tyrosine kinases, including those responsible for phosphorylation of focal adhesion proteins, and may therefore function as natural inhibitors of transformation. TC-PTP, and especially the truncated form of this enzyme (Cool et al., Proc. Natl. Acad. Sci. U.S.A. 87:7280-7284(1990)), can inhibit the transforming activity of v-erb and v-fms (Lammers et al., J. Biol. Chem. 268: 22456-22462 (1993); Zander et al., Oncogene 8:1175-1182 (1993)). Moreover, it was found that transformation by the oncogenic form of the HER2/neu gene was suppressed in NIH 3T3 fribroblasts overexpressing PTP1 B (Brown-Shimer et al., Cancer Res. 52: 478-482 (1992)).
The expression level of PTP1B was found to be increased in a mammary cell line transformed with neu (Zhay et al., Cancer Res. 53:2272-2278 (1993)). The intimate relationship between tyrosine kinases and PTPases in the development of cancer is further evidenced by the recent finding that PTP.epsilon. is highly expressed in murine mammary tumors in transgenic mice over-expressing c-neu and v-Ha-ras, but not c-myc or int-2 (Elson and Leder, J. Biol. Chem. 270:26116-26122 (1995)). Further, the human gene encoding PTP.gamma. was mapped to 3p21, a chromosomal region which is frequently deleted in renal and lung carcinomas (LaForgia et al., Proc. Natl. Acad. Sci. U.S.A. 88: 5036-5040 (1991)).
In this context, it seems significant that PTPases appear to be involved in controlling the growth of fibroblasts. In a recent study it was found that Swiss 3T3 cells harvested at high density contain a membrane-associated PTPase whose activity on an average is 8-fold higher than that of cells harvested at low or medium density (Pallen and Tong, Proc. Natl. Acad. Sci. U.S.A. 88: 6996-7000 (1991)). It was hypothesized by the authors that density-dependent inhibition of cell growth involves the regulated elevation of the activity of the PTPase(s) in question. In accordance with this view, a novel membrane-bound, receptor-type PTPase, DEP-1, showed enhanced (&gt;=10-fold) expression levels with increasing cell density of WI-38 human embryonic lung fibroblasts and in the AG1518 fibroblast cell line (Ostman et al., Proc. Natl. Acad. Sci. U.S.A. 91: 9680-9684 (1994)).
Two closely related receptor-type PTPases, PTP.kappa. and PTP.mu., can mediate homophilic cell-cell interaction when expressed in non-adherent insect cells, suggesting that these PTPases might have a normal physiological function in cell-to-cell signaling (Gebbink et al., J. Biol. Chem. 268: 16101-16104 (1993); Brady-Kalnay et al., J. Cell Biol. 122: 961-972 (1993); Sap et al., Mol. Cell. Biol. 14: 1-9 (1994)). Interestingly, PTP.kappa. and PTP.mu. do not interact with each other, despite their structural similarity (Zondag et al., J. Biol. Chem. 270: 14247-14250 (1995)). From the studies described above it is apparent that PTPases may play an important role in regulating normal cell growth. However, as pointed out above, recent studies indicate that PTPases may also function as positive mediators of intracellular signaling and thereby induce or enhance mitogenic responses. Increased activity of certain PTPases might therefore result in cellular transformation and tumor formation. Indeed, in one study over-expression of PTP.alpha. was found to lead to transformation of rat embryo fibroblasts (Zheng, supra). In addition, a novel PTP, SAP-1, was found to be highly expressed in pancreatic and colorectal cancer cells. SAP-1 is mapped to chromosome 19 region q13.4 and might be related to carcinoembryonic antigen mapped to 19q13.2 (Uchida et al., J. Biol. Chem. 269: 12220-12228 (1994)). Further, the dsPTPase, cdc25, dephosphorylates cdc2 at Thr14/Tyr-15 and thereby functions as positive regulator of mitosis (reviewed by Hunter, Cell 80:225-236 (1995)). Inhibitors of specific PTPases are therefore likely to be of significant therapeutic value in the treatment of certain forms of cancer.
PTPases: platelet aggregation
Recent studies indicate that PTPases are centrally involved in platelet aggregation. Agonist-induced platelet activation results in calpain-catalyzed cleavage of PTP1B with a concomitant 2-fold stimulation of PTPase activity (Frangioni et al., EMBO J. 12: 4843-4856 (1993)). The cleavage of PTP1B leads to subcellular relocation of the enzyme and correlates with the transition from reversible to irreversible platelet aggregation in platelet-rich plasma. In addition, the SH2 domain containing PTPase, PTP1C/SH-PTP1, was found to translocate to the cytoskeleton in platelets after thrombin stimulation in an aggregation-dependent manner (Li et al., FEBS Lett. 343: 89-93 (1994)).
Although some details in the above two studies were recently questioned there is over-all agreement that PTP1B and PTP1C play significant functional roles in platelet aggregation (Ezumi et al., J. Biol. Chem. 270:11927-11934 (1995)). In accordance with these observations, treatment of platelets with the PTPase inhibitor pervanadate leads to significant increase in tyrosine phosphorylation, secretion and aggregation (Pumiglia et al., Biochem. J. 286: 441-449 (1992)).
PTPases: osteoporosis
The rate of bone formation is determined by the number and the activity of osteoblasts, which in term are determined by the rate of proliferation and differentiation of osteoblas progenitor cells, respectively. Histomorphometric studies indicate that the osteoblast number is the primary determinant of the rate of bone formation in humans (Gruber et al., Mineral Electrolyte Metab. 12: 246-254 (1987); reviewed in Lau et al., Biochem. J. 257: 23-36 (1989)). Acid phosphatases/PTPases may be involved in negative regulation of osteoblast proliferation. Thus, fluoride, which has phosphatase inhibitory activity, has been found to increase spinal bone density in osteoporotics by increasing osteoblast proliferation (Lau et al., supra). Consistent with this observation, an osteoblastic acid phosphatase with PTPase activity was found to be highly sensitive to mitogenic concentrations of fluoride (Lau et al., J. Biol. Chem. 260: 4653-4660 (1985); Lau et al., J. Biol. Chem. 262: 1389-1397 (1987); Lau et al., Adv. Protein Phosphatases 4: 165-198 (1987)). Interestingly, it was recently found that the level of membrane-bound PTPase activity was increased dramatically when the osteoblast-like cell line UMR 106.06 was grown on collagen type-I matrix compared to uncoated tissue culture plates. Since a significant increase in PTPase activity was observed in density-dependent growth arrested fibroblasts (Pallen and Tong, Proc. Natl. Acad. Sci. 88: 6996-7000 (1991)), it might be speculated that the increased PTPase activity directly inhibits cell growth. The mitogenic action of fluoride and other phosphatase inhibitors (molybdate and vanadate) may thus be explained by their inhibition of acid phosphatases/PTPases that negatively regulate the cell proliferation of osteoblasts. The complex nature of the involvement of PTPases in bone formation is further suggested by the recent identification of a novel parathyroid regulated, receptor-like PTPase, OST-PTP, expressed in bone and testis (Mauro et al., J. Biol. Chem. 269: 30659-30667 (1994)). OST-PTP is up-regulated following differentiation and matrix formation of primary osteoblasts and subsequently down-regulated in the osteoblasts which are actively mineralizing bone in culture. It may be hypothesized that PTPase inhibitors may prevent differentiation via inhibition of OST-PTP or other PTPases thereby leading to continued proliferation. This would be in agreement with the above-mentioned effects of fluoride and the observation that the tyrosine phosphatase inhibitor orthovanadate appears to enhance osteoblast proliferation and matrix formation (Lau et al., Endocrinology 116: 2463-2468 (1988)). In addition, it was recently observed that vanadate, vanadyl and pervanadate all increased the growth of the osteoblast-like cell line UMR106. Vanadyl and pervanadate were stronger stimulators of cell growth than vanadate. Only vanadate was able to regulate the cell differentiation as measured by cell alkaline phosphatase activity (Cortizo et al., Mol. Cell. Biochem. 145: 97-102 (1995)).
PTPases: microorganisms
Dixon and coworkers have called attention to the fact that PTPases may be a key element in the pathogenic properties of Yersinia (reviewed in Clemens et al. Molecular Microbiology 5: 2617-2620 (1991)). This finding was rather surprising since tyrosine phosphate is thought to be absent in bacteria. The genus Yersinia comprises 3 species: Y. pestis (responsible for the bubonic plague), Y. pseudoturberculosis and Y. enterocolitica (causing enteritis and mesenteric lymphadenitis). Interestingly, a dual-specificity phosphatase, VH1, has been identified in Vaccinia virus (Guan et al., Nature 350: 359-263 (1991)). These observations indicate that PTPases may play critical roles in microbial and parasitic infections, and they further point to PTPase inhibitors as a novel, putative treatment principle of infectious diseases.